RL Scaling Scientist - Research Engineer for Frontier Training

anthropic; Greater London; Hybrid; GBP 375,000 - 640,000; Full time

Job summary

Anthropic is seeking a Research Engineer to design and run large-scale RL experiments, study scaling properties across model size, compute, and horizons, and translate findings into production training recipes. You will work at the interface of research and engineering, tackling open problems where robust results are quickly deployed to training pipelines.

The role emphasizes building benchmarks, bridging teams, and delivering measurable progress in frontier-scale AI systems, with a focus on

Qualifications

• Strong empirical research skills in reinforcement learning, large-scale ML training, or closely adjacent areas.
• Proven ability to own large experiments end-to-end, from design through interpretation.
• Proficiency in Python and experience with large-scale or distributed ML systems.
• Comfort operating at the research/systems boundary, including debugging where the two meet.
• Care about the societal impacts of AI and responsible scaling.

Responsibilities

• Design, run, and interpret large-scale RL experiments, with rigorous data interpretation.
• Investigate how RL improves as horizon, compute, and model size grow.
• Build and maintain benchmarks for long-horizon RL so progress is measurable and reproducible.
• Translate validated findings into production training recipes, ensuring robustness before shipping.
• Debug complex issues at the seam where research meets infrastructure, especially at scale.
• Partner with adjacent RL teams across research and engineering to advance the RL stack.

Skills

Reinforcement Learning

Large-scale ML

End-to-end experiments

Python

Debugging at scale

Education

Bachelor’s degree or equivalent

Job description

Anthropic is seeking a Research Engineer to design and run large-scale RL experiments, study scaling properties across model size, compute, and horizons, and translate findings into production training recipes. You will work at the interface of research and engineering, tackling open problems where robust results are quickly deployed to training pipelines.

The role emphasizes building benchmarks, bridging teams, and delivering measurable progress in frontier-scale AI systems, with a focus on